Decision Support Tool For Safe Downscaling of a Water Treatment Plant

Olar, Alexandru (2021) Decision Support Tool For Safe Downscaling of a Water Treatment Plant. Design Project, Industrial Engineering and Management.

Preview	Text Master_IEM_Design Project_2021_AOlar.pdf Download (2MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (122kB)

Abstract

Water Laboratorium Noord (WLN) is a subsidiary of the Waterbedrijf Groningen (Water Company Groningen, WCG) tasked with performing water analysis, research on water related topics and act as a consultant for water technologies. WCG own the drinking water treatment plant at De Punt which intakes water from the Drentsche Aa river and produces drinking water for the city of Groningen and the surrounding area. During periods of drought, usually during the summer months, the supply of water available in the river drops to levels close or lower than the plants capacity. Failure to react to these drops, and downscale the capacity of the plant to match the supply of the river could lead to massive damage to plant components and downtime. This project aims to develop a Machine Learning (ML) model to predict the supply the water available in the river using meteorological data. In order to achieve this, data from multiple weather stations together with data from multiple SCADA systems managed by the WCG and Water Board Hunze en Aa's (WB) will be included in the analysis. Multiple ML models, including Linear Regression (LR), Decision Trees (DT) and various ensemble type methods, such as Random Forest (RF) will be used to predict the streamflow of the river with multiple lead times. Two models were developed to predict the streamflow in 6 and 12 days. New features were generated after which Forward and Exhaustive Feature Selection was used to obtain the optimal subset of features.

Item Type:	Thesis (Design Project)
Supervisor name:	Mohebbi, M. and Jayawardhana, B.
Degree programme:	Industrial Engineering and Management
Thesis type:	Design Project
Language:	English
Date Deposited:	23 Mar 2021 13:00
Last Modified:	23 Mar 2021 13:00
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/24104

Actions (login required)

View Item